PROJECT SUMMARY Non-alcoholic fatty liver disease (NAFLD) is a major public health issue that affects millions of Americans, and that is increasing in prevalence with the global rise in obesity. Cutting edge human genetic approaches have identified natural human genetic variants associated with liver fat. Of these, associations with two genes: PNPLA3, and PPP1R3B, have been replicated in multiple studies. We have developed unique new mouse models to help us understand how increased PPP1R3B protects against fatty liver. PPP1R3B encodes a protein known to regulate liver glycogen, but which has only been connected to liver fat by genetic association: in other words, the role of PPP1R3B in liver fat metabolism is unknown. Interestingly, PPP1R3B is also genetically associated with multiple traits relevant to human metabolic health, including fasting insulin and glucose, plasma lactate, alkaline phosphatase, and plasma cholesterol (total, LDL and HDL cholesterol). All of these association signals map quite far from the end of the PPP1R3B gene, to a long non-coding RNA (lncRNA) of unknown function, LOC157273. Despite the considerable physical distance, genetic variants were found to correlate with increased liver PPP1R3B RNA. The minor allele (occurring in ~9% of Europeans) is associated with increased hepatic PPP1R3B mRNA expression and reduced liver and plasma lipids. Our preliminary data in mice strongly suggest that PPP1R3B is the causal gene: liver-specific PPP1R3B knockout mice (Ppp1r3b?hep) have increased hepatic and plasma lipids, whereas increasing Ppp1r3b levels in liver reduces hepatic and plasma lipids. We propose to elucidate the mechanisms by which hepatic PPP1R3B influences hepatic fat and plasma cholesterol, and to determine how the natural variants increase expression of the PPP1R3B gene. !